Malignant tumor metabolism including lipid sysnthesis and peroxidation differs from normal breast tissue and benign tumors. Some of these biochemical changes can be detected by proton MR spectroscopic imaging (MRSI) using lactate, choline, and unsaturated lipid as molecular markers of breast cancer for clinical diagnosis and prognosis. The proton NMR detection of metabolites in breast tissue, however, has been challenging due to the intense lipid and water signals that block the observation of low concentration of metabolites in breast tissue. The recent advances in magnetic resonance spectroscopy permits the detection of elevated choline level in malignant breast tumors, reflecting the rapid phosphblipid synthesis and high metabolic activities of the proliferating cells. We have developed the Sel-MQC MRSI methods to observe lactate and unsaturated lipid changes in clinical breast cancer, which eliminate lipid and water signals in a single scan by multiple quantum filtering. In this grant period, we will develop novel fast Sel-MQC techniques to improve the temporal resolution of the Sel-MQC MRSI for clinical breast cancer diagnosis. The following specific aims will be achieved: (1) To improve clinical MR diagnostic specificity of human breast cancer using the novel Sel-MQC sequences for simultaneous mapping of choline, lactate, and unsaturated lipid signals; (2) To develop fast 3D MRSI methods and improve the temporal resolution for quantitative Sel-MQC mapping of lactate, choline, and unsaturated lipid -in breast tissues. Clinical protocols will be developed to quantitatively evaluate lactate and choline, and unsaturated lipid as molecular markers of breast caner using the Sel-MQC methods on 3T clinical MRI scanners. We will studyapproximately 30 patients/yr for those who are undergoing breast MRI imaging and subsequent biopsy. The MRSI spectral data will be compared with the pathological diagnosis to determine the potential of the technique. This project will demonstrate the clinical effectiveness of MRSI as a non-invasive method using biomolecular markers for early detection of breast cancer with potentially high diagnostic specificity and sensitivity. If successful, the proton MRSI will also provide a method to evaluate responses to the breast cancer treatment regimens, as well as diagnostic tool for other extracranial cancers.